Systems and Methods for Determining Network Configurations Using Historical and Real-Time Network Metrics Data

1. A method for optimizing network performance, comprising: receiving data related to one or more network metrics for measuring current network traffic or determining network patterns; determining, based on the received data, a model associated with at least one of detection and prediction of one or more future network events; determining whether the received data is to be used for training or updating of the model; if the received data is to be used for training or updating of the model, then train or update the model, with the trained or updated model to be used with subsequently received data; if the received data is not to be used for training or updating of the model, then apply the data to the model; determining a configuration related to the one or more network devices based on the received data being applied to and the determined model; and configuring the one or more network devices according to the determined configuration.

2. The method of claim 1 , wherein the received data corresponds with a use case associated with the one or more network metrics.

3. The method of claim 2 , wherein the received data is used to determine a use case that comprises detecting network traffic associated with a predetermined protocol and with a predetermined physical layer based on at least one of: protocol properties, traffic patterns, and packet-rate specific metrics.

4. The method of claim 3 , wherein the model comprises a clustering model of network traffic patterns associated with the network traffic, the clustering model being generated based on at least one of: sources of the network traffic, applications associated with the network traffic, time periods associated with the network traffic, types of contents associated with the network traffic, and types of users associated with the network traffic.

5. The method of claim 2 , wherein the received data is used to determine a use case that comprises: supporting dynamically allocating network resources; wherein the configuration is determined based on the current network traffic and predicted demand indicated by the model.

6. The method of claim 2 , wherein the received data is used to determine a use case that comprises: adapting a traffic profile for network traffic; wherein the configuration is determined based on a predicted network load indicated by the model and comprises at least one of: a maximum segment size (MSS) for data packets, a value of maximum burst size of the network traffic, and a buffer size of a queue for the data packets.

7. The method of claim 2 , wherein the received data is used to determine a use case that comprises: determining at least one of a caching or a compression policy for webpage data; wherein the configuration is determined based on at least one of: current network load, predicted network load indicated by the model, and predicted pattern of access to a website indicated by the model.

8. The method of claim 2 , wherein the received data is used to determine a use case that comprises: detecting a malicious attack over a network; wherein the configuration is determined based on comparing a current pattern of network traffic against a historic pattern of the network traffic indicated by the model and comprises rerouting data packets away from a target device.

9. The method of claim 1 , wherein the model comprises at least one of: a linear regression model for predicting future network traffic volume or for identifying attack trends, a classification model for classifying network traffic patterns based on predetermined rules, a clustering model that groups at least one of network traffic patterns and network resources usage into clusters based on predetermined criteria, and an anomaly detection model for identifying spikes in the network traffic.

10. The method of claim 1 , wherein the generating or updating the model comprises removing noise data from the data related to one or more network metrics for measuring current network traffic; wherein the removing noise data comprises: applying the model to related data to obtain a result, and detecting non-uniformities based on the result.

11. The method of claim 1 , further comprising: determining at least one of: probability estimates and confidence intervals for the prediction of one or more future network events based on the model; and updating the model based on the at least one of probability estimates and confidence intervals.

12. An apparatus for optimizing network performance, comprising: a memory that stores a set of instructions; and a hardware processor configured to execute the set of instructions to perform: receiving data related to one or more network metrics for measuring current network traffic or determining network patterns; determining, based on the received data, a model associated with at least one of detection and prediction of one or more future network events; determining whether the received data is to be used for training or updating of the model; if the received data is to be used for training or updating of the model, then train or update the model, with the trained or updated model to be used with subsequently received data; if the received data is not to be used for training or updating of the model, then apply the data to the model; determining a configuration related to the one or more network devices based on the received data being applied to the determined model; and configuring the one or more network devices according to the determined configuration.

13. The apparatus of claim 12 , wherein the received data corresponds with a use case associated with the one or more network metrics.

14. The apparatus of claim 13 , wherein the received data is used to determine a use case that comprises: detecting network traffics associated with a predetermined protocol and with a predetermined physical layer based on at least one of: protocol properties, traffic patterns, and packet-rate specific metrics.

15. The apparatus of claim 14 , wherein the model comprises a clustering model of network traffic patterns associated with the network traffic, the clustering model being generated based on at least one of: sources of the network traffic, applications associated with the network traffic, time periods associated with the network traffic, types of contents associated with the network traffic, and types of users associated with the network traffic.

16. The apparatus of claim 13 , wherein the received data is used to determine a use case that comprises: supporting dynamically allocating network resources; wherein the configuration is determined based on the current network traffic and predicted demand indicated by the model.

17. The apparatus of claim 13 , wherein the received data is used to determine a use case that comprises: adapting a traffic profile for network traffic; wherein the configuration is determined based on a predicted network load indicated by the model and comprises at least one of: a maximum segment size (MSS) for data packets, a value of maximum burst size of the network traffic, and a buffer size of a queue for the data packets.

18. The apparatus of claim 13 , wherein the received data is used to determine a use case that comprises: determining at least one of a caching or a compression policy for webpage data; wherein the configuration is determined based on at least one of: current network load, predicted network load indicated by the model, and predicted pattern of access to a website indicated by the model.

19. The apparatus of claim 13 , wherein the received data is used to determine a use case that comprises: detecting a malicious attack over a network; wherein the configuration is determined based on comparing a current pattern of network traffic against a historic pattern of the network traffic indicated by the model and comprises rerouting data packets away from a target device.

20. The apparatus of claim 12 , wherein the model comprises at least one of: a linear regression model for predicting future network traffic volume or for identifying attack trends, a classification model for classifying network traffic patterns based on predetermined rules, a clustering model that groups at least one of network traffic patterns and network resources usage into clusters based on predetermined criteria, and an anomaly detection model for identifying spikes in network traffic.

21. The apparatus of claim 12 , wherein the generating or updating the model comprises the hardware processor being configured to execute the set of instructions to perform: removing noise data from the data related to one or more network metrics for measuring current network traffic; wherein the removing noise data comprises the hardware processor being configured to execute the set of instructions to perform: applying the model to related data to obtain a result, and detecting non-uniformities based on the result.

22. The apparatus of claim 12 , wherein the hardware processor is configured to execute the set of instructions to further perform: determining at least one of: probability estimates and confidence intervals for the prediction of one or more future network events based on the model; and updating the model based on the at least one of probability estimates and confidence intervals.

23. A non-transitory computer readable medium that stores a set of instructions that is executable by at least one processor of an apparatus to cause the apparatus to perform a method for optimizing network performance, the method comprising: receiving data related to one or more network metrics for measuring current network traffic or determining network patterns; determining, based on the received data, a model associated with at least one of detection and prediction of one or more future network events; determining whether the received data is to be used for training or updating of the model; if the received data is to be used for training or updating of the model, then train or update the model, with the trained or updated model to be used with subsequently received data; if the received data is not to be used for training or updating of the model, then apply the data to the model; determining a configuration related to the one or more network devices based on the received data being applied to the determined model; and configuring the one or more network devices according to the determined configuration.

24. The non-transitory computer readable medium of claim 23 , wherein the received data corresponds with a use case associated with the one or more network metrics.

25. The non-transitory computer readable medium of claim 24 , wherein the received data is used to determine a use case that comprises: detecting network traffic associated with a predetermined protocol and with a predetermined physical layer based on at least one of: protocol properties, traffic patterns, and packet-rate specific metrics.

26. The non-transitory computer readable medium of claim 25 , wherein the model comprises a clustering model of network traffic patterns associated with the network traffic, the clustering model being generated based on at least one of: sources of the network traffic, applications associated with the network traffic, time periods associated with the network traffic, types of contents associated with the network traffic, and types of users associated with the network traffic.

27. The non-transitory computer readable medium of claim 24 , wherein the received data is used to determine a use case that comprises: supporting dynamically allocating network resources; wherein the configuration is determined based on the current network traffic and predicted demand indicated by the model.

28. The non-transitory computer readable medium of claim 24 , wherein the received data is used to determine a use case that comprises: adapting a traffic profile for network traffic; wherein the configuration is determined based on a predicted network load indicated by the model and comprises at least one of: a maximum segment size (MSS) for data packets, a value of maximum burst size of the network traffic, and a buffer size of a queue for the data packets.

29. The non-transitory computer readable method of claim 24 , wherein the received data is used to determine a use case that comprises: determining at least one of a caching or a compression policy for webpage data; wherein the configuration is determined based on at least one of: current network load, predicted network load indicated by the model, and predicted pattern of access to a website indicated by the model.

30. The non-transitory computer readable medium of claim 24 , wherein the received data is used to determine a use case that comprises: detecting a malicious attack over a network; wherein the configuration is determined based on comparing a current pattern of network traffic against a historic pattern of the network traffic indicated by the model and comprises rerouting data packets away from a target device.

31. The non-transitory computer readable medium of claim 23 , wherein the model comprises at least one of: a linear regression model for predicting future network traffic volume or for identifying attack trends, a classification model for classifying network traffic patterns based on predetermined rules, a clustering model that groups at least one of network traffic patterns and network resources usages into clusters based on predetermined criteria, and an anomaly detection model for identifying spikes in network traffic.

32. The non-transitory computer readable medium of claim 23 , wherein the generating or updating the model comprises the non-transitory computer readable medium storing the set of instructions that is executable by at least one processor of an apparatus to cause the apparatus to further perform: removing noise data from the data related to one or more network metrics for measuring current network traffic; wherein the removing noise data comprises the non-transitory computer readable medium storing the set of instructions that is executable by at least one processor of an apparatus to cause the apparatus to further perform: applying the model to related data to obtain a result, and detecting non-uniformities based on the result.

33. The non-transitory computer readable medium of claim 32 , further storing the set of instructions that is executable by at least one processor of an apparatus to cause the apparatus to further perform: determining at least one of: probability estimates and confidence intervals for the prediction of one or more future network events based on the model; and updating the model based on the at least one of probability estimates and confidence intervals.